Method and system for managing cpu power states for heterogeneous virtualised workloads

Abstract

A method manages virtual network function components (VNFCs) in a compute infrastructure of a virtualized environment. The method includes performing, by an orchestrator system of the virtualized environment, power state-based resource pooling in the compute infrastructure based on central processing unit (CPU) power state management policies; and scheduling, by a scheduler, virtual CPUs (vCPUs) of the VNFCs on physical cores of appropriate resource pools whose CPU power states match with power profile characteristics of the vCPUs.

1 : A method for managing virtual network function components (VNFCs) in a compute infrastructure of a virtualized environment, the method comprising:
 performing, by an orchestrator system of the virtualized environment, power state-based resource pooling in the compute infrastructure based on central processing unit (CPU) power state management policies; and   scheduling, by a scheduler, virtual CPUs (vCPUs) of the VNFCs on physical cores of appropriate resource pools whose CPU power states match with power profile characteristics of the vCPUs.   
     
     
         2 : The method according to  claim 1 , further comprising:
 by a power state controller, controlling and dynamically adjusting the CPU power states of the physical cores of a resource pool at runtime based on real-time utilization metrics.   
     
     
         3 : The method according to  claim 1 , wherein the power state-based resource pooling is performed by dividing compute resources of the compute infrastructure into various power state aware zones, wherein the zones comprise at least a static P-zone with preset and static power states of the associated compute resources and a dynamic P-zone that allows runtime management of the power states of the associated compute resources. 
     
     
         4 : The method according to  claim 3 , wherein the power state-based resource pooling is performed by further dividing the associated compute resources belonging to the dynamic P-zone into power state-based groups, wherein the groups are based on operating P-state limits of the physical cores of the respective compute resources. 
     
     
         5 : The method according to  claim 1 , further comprising:
 maintaining, for each physical core, CPU power characteristics comprising operating P-state limits, in a memory component;   periodically updating the CPU power characteristics to account for runtime dynamic changes; and   using, by the scheduler, a most recent state of the memory component to schedule the vCPUs of the VNFCs on the physical cores by matching power profile requirements.   
     
     
         6 : The method according to  claim 2 , further comprising, by the power state controller:
 setting a P-state for each physical core according to a utilization for each physical core on a core level, or   associating the physical cores to their power state-based groups based on workload demand.   
     
     
         7 : The method according to  claim 1 , further comprising:
 exposing at runtime power states from multiple levels of a compute infrastructure hierarchy to the orchestrator system of the virtualized environment, wherein the multiple levels comprise at least one of a core level, node level, cluster level, PoP level and cloud instance level.   
     
     
         8 : The method according to  claim 1 , further comprising:
 using, by the orchestrator system of the virtualized environment, information about the power profile characteristics of the vCPUs provided in a VNF descriptor, to perform power state aware orchestration of the vCPUs of the VNFCs.   
     
     
         9 : The method according to  claim 1 , further comprising:
 obtaining, by a monitoring system, data on real-time and historic utilization of compute resources on at least one of a per-cloud/per-cluster/per-node/per-core basis to get power-state signatures associated with different VNFCs of the VNFCs; and   using the obtained data as training data for training an artificial intelligence/machine learning (AI/ML) analysis tool to generate power saving policies to be applied at multiple levels in a compute infrastructure hierarchy.   
     
     
         10 : The method according to  claim 1 , wherein the orchestrator system of the virtualized environment manages multiple compute infrastructures via a plurality of Points of Presence (PoP) supporting different virtualisation technologies. 
     
     
         11 : A system for managing virtual network function components (VNFCs) in a compute infrastructure of a virtualized environment, the system comprising:
 an orchestrator system configured to perform power state-based resource pooling in the compute infrastructure based on central processing unit (CPU) power state management policies; and   a scheduler configured to schedule virtual CPUs (vCPUs) of the VNFCs on physical cores of appropriate resource pools whose CPU power states match with power profile characteristics of the vCPUs.   
     
     
         12 : The system according to  claim 11 , further comprising a power state controller configured to control and dynamically adjust the CPU power states of the physical cores of a resource pool at runtime based on real-time utilization metrics. 
     
     
         13 : The system according to  claim 11 , further comprising:
 a monitoring system configured to obtain data on real-time and historic utilization of the compute resources on at least one of a per-cloud/per-cluster/per-node/per-core basis to get power-state signatures associated with different VNFCs of the VNFCs; and   an artificial intelligence/machine learning (AI/ML) analysis tool configured to use the data obtained by the monitoring system as training data to generate power saving policies to be applied at multiple levels in a compute infrastructure hierarchy.   
     
     
         14 : The system according to  claim 11 , wherein the orchestrator system comprises a network functions virtualization management and orchestration (NFV-MANO) or a service and management orchestrator (SMO) of an open radio access network (O-RAN). 
     
     
         15 : The system according to  claim 11 , wherein the VNFCs are components of a containerized virtual network function (VNF) wherein the containerized VNFs are deployed via containerized infrastructure service (CIS) clusters, wherein a CIS manager orchestrates the containerized VNFCs in the form of pods on worker CIS cluster nodes.